Beruflich Dokumente
Kultur Dokumente
http://dx.doi.org/10.1016/j.jdiacomp.2017.03.011
1056-8727/ 2017 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
L. Ji et al. / Journal of Diabetes and Its Complications 31 (2017) 11881196 1189
a r t i c l e i n f o a b s t r a c t
Article history: Aim: Contemporary global real-world data on the management of type 2 diabetes are scarce. The global DISCOVER
Received 22 December 2016 study program aims to describe the disease management patterns and a broad range of associated outcomes in
Received in revised form 23 March 2017 patients with type 2 diabetes initiating a second-line glucose-lowering therapy in routine clinical practice.
Accepted 25 March 2017 Methods: The DISCOVER program comprises two longitudinal observational studies involving more than 15,000
Available online 5 April 2017
patients in 38 countries across six continents. Study sites have been selected to be representative of type 2 diabetes
management in each country. Data will be collected at baseline (initiation of second-line therapy), at 6 months, and
Keywords:
Type 2 diabetes
yearly during a 3-year follow-up period.
Longitudinal observational study Results: The DISCOVER program will record patient, healthcare provider, and healthcare system characteristics,
Treatment patterns treatment patterns, and factors inuencing changes in therapy. In addition, disease control (e.g. achievement of
Second-line therapy glycated hemoglobin target), management of associated risk factors (e.g. hypercholesterolemia and hypertension),
Outcomes and healthcare resource utilization will be recorded. Microvascular and macrovascular complications, incidence of
Real-world evidence hypoglycemic events, and patient-reported outcomes will also be captured.
Conclusions: The DISCOVER program will provide insights into the current management of patients with type 2
diabetes worldwide, which will contribute to informing future clinical guidelines and improving patient care.
2017 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Algeria 294 Czech Republic 461 Lebanon 371 South Africa 518
Argentina 302 Denmark 42 Malaysia 342 South Korea 250
Australia 175 Egypt 584 Mexico 473 Spain 233
Austria 209 France 267 Netherlands 172 Sweden 237
Bahrain 70 India 3150 Norway 80 Taiwan 265
Brazil 444 Indonesia 238 Oman 31 Tunisia 218
Canada 386 Italy 378 Panama 92 Turkey 536
China 1298 Japan 1915 Poland 328 United Arab Emirates 102
Colombia 205 Jordan 275 Russia 601
Costa Rica 128 Kuwait 51 Saudi Arabia 585
Fig. 1. Countries participating in the DISCOVER study program (indicated in dark gray), and number of enrolled patients in each country.
system. Data will be saved immediately to a central database, and all healthcare resource utilization. In addition, information related to
eCRFs will be checked to ensure that they are completed appropri- hospitalizations or emergency department visits during follow-up
ately. Where technically feasible and allowed by local regulations, the (including reason for and duration of hospitalization, details about the
information captured in the eCRF will be linked with other sources of ward where the patient was treated, and interventions received) will
health information, such as existing electronic medical records be recorded in specic elds of the eCRF (Table 2). The investigators
(EMRs) and disease registries, to build a comprehensive data resource will be asked to contact the physician(s) in charge of the patient
for research. during hospitalization to obtain the required information. Events
Baseline data will be collected at the routine visit, during which the identied through this procedure will be classied without further
second-line glucose-lowering therapy will be prescribed. If this is not assessment.
possible (e.g. because of time constraints), the investigator will In all countries except Japan, additional patients could be enrolled
arrange for another visit to take place within 2 weeks to collect all retrospectively to compensate for short recruitment periods in
necessary information. This will include demographic and anthropo- countries where protocol approval was delayed. All eligible patients
metric data, available laboratory test results, medical history of type 2 who initiated their second-line glucose-lowering therapy between
diabetes, presence of comorbidities, co-medications, level of care, December 31, 2014 and the end of the recruitment period in each
previous glucose-lowering treatment (rst-line therapy) and reason country could be invited to participate in the study and enroll after
for change, second-line glucose-lowering drug class, HbA1c target, and providing informed consent. Therefore, for countries in which
PROs (Table 2). recruitment was closed in June 2016, the maximal length of time
During the 3-year follow-up, data will be captured at 6, 12, 24, and between initiation of second-line therapy and enrollment could be
36 months within a 4-month window (2 months) (Fig. 2A) to 18 months. For these patients, data from the baseline visit and
increase the probability that the data collection time point coincides subsequent visits that occurred in the past were collected retrospec-
with a routine patient visit; the protocol does not mandate any tively by the investigators from medical records (Fig. 2B); PRO
compulsory follow-up visits in order to ensure that the study results questionnaires were not completed for these time points.
reect routine clinical practice as much as possible. Variables to be
recorded will include physiological parameters, laboratory test 2.4.2. Patient-reported outcomes
results, change(s) in glucose-lowering therapy and reason(s) for PROs will be collected using up to four self-administered
change(s) (e.g. suboptimal glycemic control and adverse events), questionnaires depending on availability in local languages: (1) the
HbA1c level at time of change(s), occurrence of microvascular and 36-item Short-form Health Survey version 2 (SF-36v2) will be used to
macrovascular events, occurrence of minor and major hypoglycemic measure general health status across eight domains: vitality, physical
events, changes in comorbidities and co-medications, PROs, and functioning, bodily pain, general health perceptions, physical role
1192 L. Ji et al. / Journal of Diabetes and Its Complications 31 (2017) 11881196
Fig. 2. Study timelines. A. Patients enrolled on the day of initiation of second-line glucose-lowering therapy. B. Patients enrolled retrospectively (after initiation of second-line
glucose-lowering therapy).
functioning, emotional role functioning, social role functioning, and data from mandatory national health registries (e.g. information on
mental health; 20 (2) the revised Hypoglycemia Fear Survey (HFS-II), a hospitalizations, prescribed drugs, and cause of death) to create a
33-item survey using ve-point Likert scales (score range: 0132), de-identied study database.
will be used to assess behaviors and worries relating to fear of A similar approach will be used in Canada. All patients were
hypoglycemia; 21 (3) a seven-item questionnaire will be used to assess identied retrospectively, using the IMS Brogan EMR database. Eligible
patients' lifestyles as unhealthy, intermediate, healthy, or very patients were invited to participate and enrolled after providing
healthy, based on questions relating to smoking, alcohol intake, informed consent. Most data will be extracted from the EMR database
physical activity, and diet; 22 and (4) a two-item questionnaire will be for all time points. A short eCRF completed by the patient will also be
used to determine whether patients avoided healthcare and/or used to collect additional data not routinely captured in the EMR
medication because of cost. In addition to the SF-36v2 questionnaire, database (e.g. hypoglycemic events and PROs). The eCRF was not used at
the J-DISCOVER study will use slightly different questionnaires: (1) baseline because patients were enrolled retrospectively. This approach
the Japanese version of the diabetes treatment satisfaction question- will provide a more limited dataset than in other DISCOVER countries;
naire (DTSQ), an 8-item survey using 7-point Likert scales, will be therefore, data from Canada will only be included in the global analysis
used to assess treatment satisfaction and measure the perceived of the DISCOVER results when appropriate.
frequency of hyperglycemia and hypoglycemia; 23 (2) the brief In Canada, Denmark, France, Norway, and Sweden, it will also be
self-administered diet-history questionnaire (BDHQ) will be used to possible to collect EMR data beyond the 3-year follow-up period,
assess dietary habits; 24,25 and (3) the short version of the Interna- providing insights on longer-term disease management and out-
tional Physical Activity Questionnaire (IPAQ-SV), a 9-item survey, will comes. In these ve countries, retrospective data will be collected
be used to estimate physical activity levels. 26,27 from EMRs and/or national health registries for all patients with type
2 diabetes treated at each primary care site participating in the
2.4.3. Data collection from electronic medical records DISCOVER study (including patients not enrolled in the prospective
In Denmark, France, Norway, and Sweden, available study DISCOVER study) and analyzed separately. This approach will enable a
variables will be extracted automatically from existing primary care longer observation period in a large cohort and provide important
EMR databases. For all patients meeting the inclusion criteria at each data to assess the external validity of the results from the global
participating site, EMR data will be extracted at baseline and yearly DISCOVER study. The possibility of implementing a similar extended
thereafter. At the time of de-identied EMR data extraction, a key follow-up in other DISCOVER countries will be explored.
code will be created that links each patient identication number to
their registry data. This key code will be encrypted, password
protected, and stored separately from the database used for analysis. 2.4.4. Adverse event reporting
Data not routinely captured in EMR databases (e.g. reason for DISCOVER is a non-interventional study program and there is no
treatment change) will be collected by investigators using an requirement to report adverse events. Adverse drug reactions
additional questionnaire. These data will then be linked back to the observed in patients participating in DISCOVER will be reported to
EMR to create a full dataset for each patient. In addition, in Denmark, health authorities as required by local regulations and/or if the
Sweden, and Norway, data extracted from EMRs will be merged with investigator considers reporting to be appropriate.
L. Ji et al. / Journal of Diabetes and Its Complications 31 (2017) 11881196 1193
Table 2
Data collection at baseline (initiation of second-line therapy) and 6, 12, 24, and 36 months.
HbA1c: glycated hemoglobin, HDL-C: high-density lipoprotein cholesterol, HFS-II: revised Hypoglycemia Fear Survey, LDL-C: low-density lipoprotein cholesterol, PCP: primary care
practitioner, SF-36v2: 36-Item Short-Form Health Survey version 2.
a
Follow-up visits at 6, 12, 24, and 36 months within a 4-month window ( 2 months).
b
Patient-reported outcomes will not be collected for baseline and for any data collection time point occurring before enrollment for patients identied retrospectively. Patients
will only complete questionnaires available in their country.
1194 L. Ji et al. / Journal of Diabetes and Its Complications 31 (2017) 11881196
2.5. Statistical analysis neous population. The study will provide a comprehensive overview
of current clinical practices in the treatment of patients with type 2
Descriptive statistics will be used to summarize demographic diabetes after failure of rst-line oral therapy across 38 countries, and
variables, patient characteristics, treatment patterns, changes in will identify potential differences between practice and clinical
HbA1c level, blood glucose, lipid prole, body weight, body mass recommendations. The study is designed to generate sufcient data
index, and blood pressure, incidence of complications and hypogly- to analyze determinants of treatment decisions, and associations
cemic events, PROs, and healthcare resource utilization. Data will be between treatment patterns and several disease-specic outcomes,
stratied by pharmacological drug class and/or individual agent when including glycemic control, microvascular and macrovascular events,
clinically relevant. Hierarchical multivariate regression analyses and hypoglycemic episodes, and PROs. In addition, the large sample size
general linear models will be used to determine potential predictors will allow subanalyses to help to identify patient proles or subgroups
of treatment choices and clinical outcomes. When applicable, that respond well to specic therapeutic approaches. Importantly, the
model-based point estimates of odds ratios or risk ratios, as DISCOVER study program will generate real-world data on the
appropriate, corresponding 95% condence intervals, and p values management of type 2 diabetes and associated outcomes in countries
will be reported. For analyses of clinical outcomes during follow-up, with alarmingly high and rising prevalences of type 2 diabetes that
Cox models and mixed models for longitudinal data will be used, have been under-studied to date.
adjusted for age and sex, and using country as the indicator variable. DISCOVER is primarily a descriptive study, and representativeness
Time-to-event analyses will be used to describe treatment patterns of participating patients and physicians is key to meeting the study
(e.g. initiation of third-line therapy). objectives and ensuring external validity of the results. Physicians
The minimum sample size was estimated to be 11,100 patients, invited to participate were therefore carefully selected across
based on the following criteria: (1) the intention to have a minimum different specialties, care settings, and geographic regions in order
of 200 patients in any given group of patients to be analyzed (e.g. to ensure that the ndings would be representative of the manage-
patients receiving a certain drug class or patients from one country), ment of patients with type 2 diabetes in each country. Nevertheless,
which would result in a precision of at least 7% for any expected physician selection bias should be considered when interpreting the
qualitative variable at a frequency of 5%95%; (2) at least 200 patients results. Physicians who are more involved and/or competent in the
meeting each of the composite endpoints of macrovascular compli- treatment and management of diabetes may be more likely to
cations (cardiovascular death, myocardial infarction, ischemic stroke, participate and to treat patients earlier. In this case, the results of
hospitalization for heart failure, coronary revascularization) and DISCOVER may slightly overestimate disease control and quality of
microvascular complications (neuropathy, retinopathy, nephropathy) care in day-to-day clinical practice. However, the large number of
at years 1, 2, and 3, based on expected yearly rates for these events participating investigators (N 1000) across all specialties and from
(5.5% for the composite of macrovascular complications; 10.4% for the different clinical settings, and the inclusion of a diverse patient
composite of microvascular complications); (3) at least 200 patients population, are likely to provide an accurate overview of diabetes
meeting each individual macrovascular and microvascular endpoint management in each participating country or region.
at year 3 (expected rates over 3 years: 2.4%4.6% for macrovascular DISCOVER uses PRO questionnaires and therefore relies on
complications; 3.5%15.7% for microvascular complications); and (4) patients' recollection to determine the incidence of some events
an estimated attrition rate of 15% per year of follow-up. (e.g. minor hypoglycemic events); recall bias may therefore also
All statistical analyses will be performed using the SAS statistical inuence the results of the study. However, this potential misclassi-
software system (SAS Institute, Inc., Cary, NC). cation of outcomes is unlikely to be systematically associated with
patient characteristics, risk of complications, or exposure to
2.6. Funding and responsibilities glucose-lowering agents. In addition, results from these question-
naires should be interpreted cautiously in light of data missing from
DISCOVER is funded by AstraZeneca. As it is a non-interventional patients enrolled retrospectively. Loss to follow-up might also be a
study, no drugs are supplied or funded. The DISCOVER Scientic source of bias. Patients with poorly controlled diabetes may be more
Committee comprises 12 scientists (L.J., F.B., B.C., M.B.G., M.K., K.K., likely to drop out over the 3-year follow-up than those with better
A.N., S.P., W.R., M.V.S., I.S., and H.W.) and the following AstraZeneca outcomes, which could result in an underestimation of negative
members: the Scientic Project Leader (J.M.), the Global Medical outcomes. It should also be noted that patient visits are not
Affairs Leader Diabetes (P.F.), the Global Medical Affairs Medical determined by the protocol; therefore, data collection during
Evidence and Observational Research Diabetes Lead (N.H.), the follow-up will rely on patients attending regular routine visits (the
Medical Affairs Senior Director Japan (K.H.), and former International frequency of which may be country dependent) and on investigators
Region Medical Directors (F.S. and G.M.). being able to collect all relevant information (e.g. data about
hospitalizations or emergency department visits). Finally, in common
3. Discussion with all observational studies, confounding should be considered
when interpreting the results. However, the available variables
Although large, international, non-interventional studies of patients collected using the study eCRF will be used in multivariate analyses
with type 2 diabetes have been conducted,2841 they only examined to minimize confounding when assessing associations between
patients treated with insulin 2931,34,36,37,3941 or another specic clinical outcomes and patient, physician, and healthcare system
glucose-lowering drug class,32,33,35,36,38 were limited to ve or fewer characteristics, and treatment patterns.
countries29,34,37 or included a relatively small number of patients The DISCOVER study program also has major strengths in addition
(b 1000),37 were of short duration (follow-up 6 months),29,3133,3841 to the large sample size and the broad geographical representation.
or were cross-sectional.28 In addition, many of these studies focused on First, the naturalistic design of the studies, whereby the chances of
drug safety and/or efcacy29,3133,35,3840 and few assessed quality of protocol-induced ndings and protocol-induced healthcare resource
life, 28,39,41 incidence of hypoglycemic events, 29,31,41 or treatment use are minimized, will help to ensure external validity of the results.
patterns.30,31,34,36,37,42 In addition, the same eCRF will be used in all countries, ensuring
DISCOVER is a global, prospective, observational study of patients consistent data collection. This user-friendly eCRF, combined with a
with type 2 diabetes, and it will be the rst to report treatment relatively small target number of patients (b 20 on average) per
patterns after initiation of second-line therapy regardless of the investigator, will ensure that the time burden for investigators is
agent(s) prescribed, in a large, contemporary, diverse, and heteroge- minimal and that data are collected completely and accurately for all
L. Ji et al. / Journal of Diabetes and Its Complications 31 (2017) 11881196 1195
patients. The long follow-up period is also a key strength of DISCOVER 7. Preis SR, Pencina MJ, Hwang SJ, D'Agostino Sr RB, Savage PJ, Levy D, et al. Trends in
cardiovascular disease risk factors in individuals with and without diabetes
and will allow the study of treatment patterns beyond initiation of mellitus in the Framingham heart study. Circulation. 2009;120:212-20.
second-line therapy. In some countries, prospective data collection 8. Chinese Diabetes Society. Chinese guideline for type 2 diabetes prevention (2013).
will be complemented by data extracted from existing EMR databases, Chinese J Diabetes. 2014;22:2-42.
9. Garber AJ, Abrahamson MJ, Barzilay JI, Blonde L, Bloomgarden ZT, Bush MA, et al.
thus providing the opportunity to follow up some patients for longer AACE/ACE comprehensive diabetes management algorithm 2015. Endocr Pract.
than 3 years. Comparison of data extracted from EMRs and data 2015;21:438-47.
collected with the eCRF may also inform the design of future 10. International Diabetes Federation. Global Guideline for Type 2 Diabetes. Available
at http://www.idf.org/sites/default/les/IDF-Guideline-for-Type-2-Diabetes.pdf
observational studies. In addition, the DISCOVER study program will
2012. [Accessed November 2016, 30].
allow platforms for future research to be set up in countries with no 11. Inzucchi SE, Bergenstal RM, Buse JB, Diamant M, Ferrannini E, Nauck M, et al.
previous systems for the capture of observational data from patients Management of hyperglycaemia in type 2 diabetes, 2015: a patient-centred
approach. Update to a position statement of the American Diabetes Association and
with type 2 diabetes.
the European Association for the Study of Diabetes. Diabetologia. 2015;58:429-42.
It should be noted that, although the DISCOVER study program 12. Qaseem A, Humphrey LL, Sweet DE, Starkey M, Shekelle P. Oral pharmacologic
does not cover the USA, it has been developed in parallel with the treatment of type 2 diabetes mellitus: a clinical practice guideline from the
recently launched Diabetes Collaborative Registry (DCR), 43 which has American College of Physicians. Ann Intern Med. 2012;156:218-31.
13. Holman RR, Paul SK, Bethel MA, Matthews DR, Neil HA. 10-year follow-up of
the potential to become the largest cross-specialty clinical registry to intensive glucose control in type 2 diabetes. N Engl J Med. 2008;359:1577-89.
track, evaluate, and eventually improve the quality of diabetes and 14. Patel A, MacMahon S, Chalmers J, Neal B, Billot L, Woodward M, et al. Intensive
metabolic care in the USA. The DCR captures similar data via EMR data blood glucose control and vascular outcomes in patients with type 2 diabetes.
N Engl J Med. 2008;358:2560-72.
extraction and will complement the DISCOVER results from 38 other 15. Ray KK, Seshasai SR, Wijesuriya S, Sivakumaran R, Nethercott S, Preiss D, et al.
countries. Studies of this type are essential to review trends in Effect of intensive control of glucose on cardiovascular outcomes and death in
prescribing practices, diabetes management, and clinical outcomes, patients with diabetes mellitus: a meta-analysis of randomised controlled trials.
Lancet. 2009;373:1765-72.
in order to help improve the standard of care for patients with type 16. Stratton IM, Adler AI, Neil HA, Matthews DR, Manley SE, Cull CA, et al. Association of
2 diabetes. glycaemia with macrovascular and microvascular complications of type 2 diabetes
(UKPDS 35): prospective observational study. BMJ. 2000;321:405-12.
17. UK Prospective Diabetes Study (UKPDS) Group. Intensive blood-glucose control with
4. Conclusion sulphonylureas or insulin compared with conventional treatment and risk of
complications in patients with type 2 diabetes (UKPDS 33). Lancet. 1998;352:837-53.
18. Marso SP, Daniels GH, Brown-Frandsen K, Kristensen P, Mann JF, Nauck MA, et al. Liraglutide
The DISCOVER study program is a global initiative that will provide and cardiovascular outcomes in type 2 diabetes. N Engl J Med. 2016;375:311-22.
valuable insights into the treatment of patients with type 2 diabetes in 19. Zinman B, Wanner C, Lachin JM, Fitchett D, Bluhmki E, Hantel S, et al. Empagliozin,
real-world settings and will highlight differences between daily cardiovascular outcomes, and mortality in type 2 diabetes. N Engl J Med. 2015;373:
2117-28.
clinical practice and recommendations in current national and 20. Ware JE, Kosinski M, Bjorner JB, Turner-Bowker DM, Gandek B, Maruish ME. User's
international guidelines. In addition, collection of patient-centric Manual for the SF-36v2 Health Survey. Lincoln, RI: QualityMetric Incorporated. 2007.
health status and healthcare resource utilization data will enable the 21. Stargardt T, Gonder-Frederick L, Krobot KJ, Alexander CM. Fear of hypoglycaemia:
dening a minimum clinically important difference in patients with type 2
evaluation of important outcomes beyond glycemic control and
diabetes. Health Qual Life Outcomes. 2009;7:91.
clinical events. The DISCOVER study program will also allow 22. Carlsson AC, Wandell PE, Gigante B, Leander K, Hellenius ML, de Faire U. Seven
benchmarking of clinical practice between and within countries, modiable lifestyle factors predict reduced risk for ischemic cardiovascular disease
and all-cause mortality regardless of body mass index: a cohort study. Int J Cardiol.
will help to inform future updates to clinical guidelines, and will
2013;168:946-52.
further improve the care of patients with type 2 diabetes. 23. Ishi H, Bradley C, Riazi A, Barendse S, Yamamoyo T. Tonyobyo ticho manzokudo
Supplementary data to this article can be found online at http://dx. sitsumonhyo (DTSQ) no nihongo honyaku to hyoka ni kansuru kenkyu [the
doi.org/10.1016/j.jdiacomp.2017.03.011. Japanese version of the diabetes treatment satisfaction questionnaire (DSTQ):
translation and clinical evaluation]. Igaku no Ayumi. 2000;1992:809-14.
24. Kobayashi S, Murakami K, Sasaki S, Okubo H, Hirota N, Notsu A, et al. Comparison of
relative validity of food group intakes estimated by comprehensive and brief-type
Funding self-administered diet history questionnaires against 16 d dietary records in
Japanese adults. Public Health Nutr. 2011;14:1200-11.
The DISCOVER study is funded by AstraZeneca. 25. Sasaki S, Yanagibori R, Amano K. Self-administered diet history questionnaire
developed for health education: a relative validation of the test-version by
comparison with 3-day diet record in women. J Epilepsy. 1998;8:203-15.
Acknowledgments 26. Craig CL, Marshall AL, Sjostrom M, Bauman AE, Booth ML, Ainsworth BE, et al.
International physical activity questionnaire: 12-country reliability and validity.
Med Sci Sports Exerc. 2003;35:1381-95.
The authors would like to thank all investigators and patients 27. Murase N, Katsumura T, Ueda C, Inoue S, Shimomitsu T. Validity and reliability of
participating in the DISCOVER study program. Medical writing Japanese version of international physical activity questionnaire. J Health Welf Stat.
2002;49:1-9.
support was provided by Stphane Pintat PhD of Oxford PharmaGen-
28. Bradley C, de Pablos-Velasco P, Parhofer KG, Eschwege E, Gonder-Frederick L,
esis, Oxford, UK, and was funded by AstraZeneca. Simon D. PANORAMA: a European study to evaluate quality of life and treatment
satisfaction in patients with type-2 diabetes mellitus study design. Prim Care
Diabetes. 2011;5:231-9.
References 29. Echtay A, Tsur A, Hasan MI, Abu-Hijleh MO, Al Khatib N, Andari E, et al. Clinical
experience with insulin detemir in patients with type 2 diabetes from the near east
1. International Diabetes Federation. IDF Diabetes Atlas7th ed. 2015. [Available at: countries. Diabetes Ther. 2013;4:399-408.
http://www.diabetesatlas.org/; Accessed 10 February 2017]. 30. Freemantle N, Balkau B, Danchin N, Wang E, Marre M, Vespasiani G, et al. Factors
2. Boyle JP, Engelgau MM, Thompson TJ, Goldschmid MG, Beckles GL, Timberlake DS, inuencing initial choice of insulin therapy in a large international non-interventional
et al. Estimating prevalence of type 1 and type 2 diabetes in a population of African study of people with type 2 diabetes. Diabetes Obes Metab. 2012;14:901-9.
Americans with diabetes mellitus. Am J Epidemiol. 1999;149:55-63. 31. Khunti K, Caputo S, Damci T, Dzida GJ, Ji Q, Kaiser M, et al. The safety and efcacy of adding
3. Bruno G, Runzo C, Cavallo-Perin P, Merletti F, Rivetti M, Pinach S, et al. Incidence of once-daily insulin detemir to oral hypoglycaemic agents in patients with type 2 diabetes
type 1 and type 2 diabetes in adults aged 30-49 years: the population-based in a clinical practice setting in 10 countries. Diabetes Obes Metab. 2012;14:1129-36.
registry in the province of Turin, Italy. Diabetes Care. 2005;28:2613-9. 32. Kim CH, Han KA, Oh HJ, Tan KE, Sothiratnam R, Tjokroprawiro A, et al. Safety,
4. Evans JM, Newton RW, Ruta DA, MacDonald TM, Morris AD. Socio-economic status, tolerability, and efcacy of metformin extended-release oral antidiabetic therapy in
obesity and prevalence of type 1 and type 2 diabetes mellitus. Diabet Med. 2000;17: patients with type 2 diabetes: an observational trial in Asia. J Diabetes. 2012;4:395-406.
478-80. 33. Li C, Hung YJ, Qamruddin K, Aziz MF, Stein H, Schmidt B. International
5. Holman N, Young B, Gadsby R. Current prevalence of type 1 and type 2 diabetes in noninterventional study of acarbose treatment in patients with type 2 diabetes
adults and children in the UK. Diabet Med. 2015;32:1119-20. mellitus. Diabetes Res Clin Pract. 2011;92:57-64.
6. Morrish NJ, Wang SL, Stevens LK, Fuller JH, Keen H. Mortality and causes of death in 34. Liebl A, Jones S, Goday A, Benroubi M, Castell C, Haupt A, et al. Clinical outcomes
the WHO multinational study of vascular disease in diabetes. Diabetologia. after insulin initiation in patients with type 2 diabetes: 24-month results from
2001;44:S14-21. INSTIGATE. Diabetes Ther. 2012;3:9.
1196 L. Ji et al. / Journal of Diabetes and Its Complications 31 (2017) 11881196
35. Mathieu C, Barnett AH, Brath H, Conget I, de Castro JJ, Goke R, et al. Effectiveness biphasic insulin analogs in daily clinical practice. Diabetes Res Clin Pract. 2010;88:
and tolerability of second-line therapy with vildagliptin vs. other oral agents in S11-6.
type 2 diabetes: a real-life worldwide observational study (EDGE). Int J Clin Pract. 40. Tsai ST, Pathan F, Ji L, Yeung VT, Chadha M, Suastika K, et al. First insulinization with
2013;67:947-56. basal insulin in patients with type 2 diabetes in a real-world setting in Asia. J
36. Matthaei S, Reaney M, Mathieu C, Ostenson CG, Krarup T, Guerci B, et al. Patients Diabetes. 2011;3:208-16.
with type 2 diabetes initiating exenatide twice daily or insulin in clinical practice: 41. Valensi P, Benroubi M, Borzi V, Gumprecht J, Kawamori R, Shaban J, et al. The
CHOICE study. Diabetes Ther. 2012;3:6. IMPROVE study a multinational, observational study in type 2 diabetes: baseline
37. Oguz A, Benroubi M, Brismar K, Melo P, Morar C, Cleall SP, et al. Clinical outcomes characteristics from eight national cohorts. Int J Clin Pract. 2008;62:1809-19.
after 24 months of insulin therapy in patients with type 2 diabetes in ve 42. Vichayanrat A, Matawaran BJ, Wibudi A, Ferdous HS, Aamir AH, Aggarwal SK, et al.
countries: results from the TREAT study. Curr Med Res Opin. 2013;29:911-20. Assessment of baseline characteristics, glycemic control and oral antidiabetic
38. Rosales R, Abou Jaoude E, Al-Arouj M, Fawwad A, Orabi A, Shah P, et al. Clinical treatment in Asian patients with diabetes: the registry for assessing OAD usage in
effectiveness and safety of vildagliptin in N19 000 patients with type 2 diabetes: diabetes management (REASON) Asia study. J Diabetes. 2013;5:309-18.
the GUARD study. Diabetes Obes Metab. 2015;17:603-7. 43. Arnold SV, Inzucchi SE, McGuire DK, Mehta SN, Goyal A, Sperling LS, et al. Evaluating
39. Shah SN, Litwak L, Haddad J, Chakkarwar PN, Hajjaji I. The A1chieve study: a 60 the quality of comprehensive cardiometabolic care for patients with type 2 diabetes in
000-person, global, prospective, observational study of basal, meal-time, and the U.S.: the diabetes collaborative registry. Diabetes Care. 2016;39:e99-101.